- Home
- About Us
- Initiatives
- News
- Resources
- Searchable Databases
- Contact Us
State Summary
Colorado's audit law was signed into law in 2005. Audits are conducted for all races, completed before certification, and audit results are binding upon the official results, though they do not lead to full recounts.
Colorado is also in the process of moving towards risk-limiting audits, to be fully implemented by 2014. In 2009, the legislature passed a law authorizing the Secretary of State to develop regulations for risk-limiting audits, to pilot such audits in select counties during the 2010 election cycle, and to aid all counties in preparing for timely implementation. For more information, see the section on Addressing Discrepancies, below.
For polling places and Early Voting Devices, it's 5% of machines, 100% of the ballots. For absentee ballots, it's at least one scanner, with 5% of the ballots (not less than 500 ballots.) If less than 500 absentee ballots are counted, than 20% of all absentee ballots are counted.
If a discrepancy is found that cannot be explained by voter error, the county clerk and recorder, in consultation with the canvassing board, must investigate and take any action deemed necessary.
The random selection of machines to test is currently scheduled to happen "within twenty-four (24) hours of the close of polls"". In the case of a delayed count or delayed reporting, this should not happen until the machines are done counting and a tally has been obtained and made known. A report of the audit must be posted within 24 hours of receipt by SOS office.
Yes.


"See also the rules from the Secretary of State, e.g. 11.5.4 Post-Election Audit. They are watered down from the law, and allow partial recounts rather than actual audits." "The sampling method includes only all ballots on 5% machines by serial number. Some machines do not provide election sub tallies... sometimes these are audited by machine recount instead or by hand count of VVPAT compared to hand count of printed cast vote records from the device (generally these are identical records) The device made by HART does not even sub tally the cast vote records- so these auditsare also not audits of the actual election tally. "Early voting is done on serial numbers which overlap with precinct voting serial numbers. These serial numbers are not repeated in the pool of serial numbers for random selection, hence early voting sub tallies are usually not audited."